Text Detection in Natural Images Based on Character Classification

Abstract

Abstract.Text information in images is very important for image understanding. In this paper, a text location method based on character classification is proposed. The and-valley image (AVI) and the and-ridge image (ARI) are first extracted from the input image. Then character components are detected from the AVI and ARI respectively, and then these components are sent to a character classifier. Finally，text region can be generated by merging all the recognized components. This approach is robust to font style, font size, font color and the background complexity. It is demonstrated in the experiments that our method is efficient.

ICDAR 2011 Robust Reading Competition Challenge 2: Reading Text in Scene Images

“…8-direction gradient direction features are extracted for character classification which is different to [24] • Group all character connected components to text regions.…”

“…In our system T area = 0.6, s1, s2 and s3 are decided according to the input image's width and height: s1=0.5, s2 = 1, s3=2, if max(width,height) > 1600 s1=1, s2=2, s3=4, else if max(width,height) > 800 s1=1, s2=2, s3=6 otherwise The detection method on each scaled image includes the following steps [24]:…”

ICDAR 2011 Robust Reading Competition - Challenge 1: Reading Text in Born-Digital Images (Web and Email)

“…In each of the scales, the And-Ridge and And-Valley images are calculated as detailed in [16]. Connected component analysis is then performed in the above images, and resulting components are subsequently classified as character and non-character.…”

An integrated artificial vision framework for assisting visually impaired users

“…For these reasons in the last decades many "smart systems" tried to address the so-called OCR in unconstrained environments (see, for instance, [47][48][49][50][51] ) in the attempt of bridging the gap between automatic document reading (which has largely been solved for printed latin characters) and a more generic automatic reading. In recent years, large improvements have been made in this direction.…”